Volume 8, September 2016, Pages 225–229

Open Access
Data Article

Fungal diversity and ecosystem function data from wine fermentation vats and microcosms

  • a Max Planck Institute for Evolutionary Biology, Plön 24306, Germany
  • b The Galton Laboratory, Department of Genetics, Evolution, and Environment, University College London, London WC1E 6BT, UK


Grape must is the precursor to wine, and consists of grape juice and its resident microbial community. We used Illumina MiSeq® to track changes in must fungal community composition over time in winery vats and laboratory microcosms. We also measured glucose consumption and biomass in microcosms derived directly from must, and glucose consumption in artificially assembled microcosms. Functional impacts of individual must yeasts in artificially assembled communities were calculated using a "keystone index," developed for “Species richness influences wine ecosystem function through a dominant species” [1]. Community composition data and functional measurements are included in this article. DNA sequences were deposited in GenBank (GenBank: SRP073276). Discussion of must succession and ecosystem functioning in must are provided in [1].

Specifications Table

Subject areaBiology
More specific subject areaEcology, Microbiology, Mycology, Oenology
Type of dataTables
How data was acquiredIllumina MiSeq® sequencing; colorimetric glucose (HK) assay kit; microbalance weighing; colony-forming unit (CFU) counting
Data formatRaw and analysed
Experimental factorsFermentation age in winery vats; inoculum dilution in microcosms; added genus in artificially assembled communities
Experimental featuresFermenting grape must from winery vats and laboratory microcosms was sequenced for fungal-specific amplicons. Glucose and biomass were measured in microcosms. Glucose was also measured in artificially assembled communities derived from individual yeast isolates and microcosms.
Data source locationSan Polino Winery, Montalcino, Italy
Data accessibilityData are within this article. Raw sequence data are available at GenBank via the accession number GenBank: SRP073276.
Full-size table

Value of the data

This dataset is one of very few must fungal datasets measured over successional time using high-throughput sequencing.

Fungal succession over time in fermenting must can be contrasted with datasets. from different winery environments or gathered using different enumeration techniques (e.g., culture-dependent enumeration).

Measured sugar utilization of yeasts and microbial communities can serve as a starting point for studies of yeast function during wine development.

The keystone index can be used to compare disproportionate functional impacts among microbes from a variety of environments.

1. Data

Data include fungal DNA amplicon sequences, OTUs, and taxonomic data from fermenting must in a winery, in a laboratory microcosm experiment, and in control communities. Associated experimental and metadata are provided in separate tables. Experimental data include microcosm biomass and glucose concentrations, plus the data needed to calculate a "keystone index," described in [1]below, for twenty microcosm yeast isolates. Metadata include winery vat identity, fermentation age, microcosm treatment, microcosm age, and microcosm replicate.

2. Experimental design, materials and methods

2.1. Fermenting grape must

All must samples were collected in October and November, 2013 from the San Polino winery in Montalcino, Italy. All winery fermentation vats were filled with must from Sangiovese grapes harvested from five vineyards, all within 5 km of the winery. Vat volumes range from 3000 to 3800 l. We collected must samples from five vats approximately every 12–24 h over 13 days starting from the day the first vat was completely filled. One ml of grape must was collected at each timepoint. To prevent further fermentation during storage and transport, we centrifuged must samples for 5 min at 6000 rpm in a tabletop microcentrifuge and fixed the pelleted cells in 250–500 μl 100% ethanol. Samples were stored at ambient temperature until DNA extraction (19 days or less), and alcohol was removed from each sample before DNA extraction. DNA was extracted using the MasterPure™ Yeast DNA Purification Kit (Epicentre, Madison, Wisconsin, USA) following the manufacturer׳s instructions.

Must samples were also collected from six vats or vat mixtures once fermentation was completed, after the winemakers had filtered the fermented must. Post-filtration samples were transported at ambient temperature without treatment for seven days before DNA extraction. The winemakers combined the contents of some vats during filtration, and two post-filtration samples were mixtures of two vats each. We assigned each of these two mixtures to the vat which contributed the most volume to the mixture (i.e., a sample consisting of 54% Vat 17 must and 46% Vat 1 must was analysed as Vat 17 and a sample consisting of 67% Vat 22 must and 33% Vat 20 must was analysed as Vat 22). The total number of must samples collected ranged from 6 to 23 per vat. Two additional vats were only sampled once, after filtration. A summary of all fermentation vat samples including fermentation age and vat identity is provided in Table 1.

2.2. Microcosm experiment

We sequenced fungal diversity and measured biomass and glucose consumption in small volumes of fermenting grape must (microcosms). We prepared ten replicates each of five dilution treatments plus uninoculated controls (Fig. S1 in [1]). Treatments included undilute unsterilized grape must and unsterilized must serially diluted 1:10, 1:103, 1:105, and 1:107 with 0.22-μm-filter-sterilized must (D0, D1, D3, D5, and D7, respectively). One millilitre was removed from each inoculated microcosm for DNA sequencing before incubation, and the remaining 10 ml microcosms were incubated for 14 days at 30 °C with 200 rpm shaking. Inoculum sizes ranged from about 50 to 5×108 colony-forming units (CFUs) per 10 ml microcosm. All must originated from a single vat (Vat 17). Must was collected 64 h after the vat was filled, and transported on ice for 24 h before microcosm preparation.

In addition to the cells harvested before incubation, we also harvested cells for DNA sequencing and measured microcosm biomasses and glucose concentrations after 14 days. Cells were harvested from all inoculated microcosms by centrifuging 1 ml of each microcosm (10 min at 16,837 rcf) and removing the supernatant. DNA was extracted as above. To measure biomass, we centrifuged a second 1 ml from each microcosm, dried each pellet at 80 °C for 38 h, and weighed pellets on a microbalance. We corrected biomass values by subtracting average uninoculated control values from each treated biomass value, but we ignored biomass data of D0 microcosms because these microcosms contained undilute grape solids. Supernatants were retained for glucose concentration assays. We decolourized supernatants by incubating 250 μl of filter-sterilized supernatant with 25–50 mg activated carbon pellets for 24 h. Glucose concentration was then measured using a Glucose (HK) Assay Kit (Sigma®, St. Louis, Missouri, USA), according to the manufacturer׳s instructions. Microcosm glucose values less than 0.14 mg/ml were assumed to be below the limit of kit detection, and were assigned a value of zero. We transformed microcosm glucose into percentage total glucose consumed by normalizing glucose concentrations to uninoculated controls. A summary of all sequenced microcosm samples is in Table 2, and glucose and biomass data are in Table 3.

2.3. Constructed control samples

Constructed control samples were known numbers of CFUs of three grape must yeasts (S. cerevisiae, Hanseniaspora uvarum, and Metschnikowia sp.) in grape must. DNA was extracted from constructed-control samples as described above for microcosm samples. CFU numbers are provided for each constructed control sample in Table 4.